Convolutional Neural Networks Do Work with Pre-Defined Filters

TL;DR

This paper introduces Pre-defined Filter CNNs (PFCNNs), which use fixed convolution kernels and learn only linear combinations, demonstrating effective feature learning despite strict constraints across multiple datasets.

Contribution

The paper proposes a novel CNN architecture with fixed pre-defined filters, challenging traditional training methods and providing new insights into feature extraction in deep networks.

Findings

PFCNNs achieve competitive accuracy on multiple datasets.

Fixed kernels can still learn complex discriminative features.

The approach offers a new perspective on information processing in CNNs.

Abstract

We present a novel class of Convolutional Neural Networks called Pre-defined Filter Convolutional Neural Networks (PFCNNs), where all nxn convolution kernels with n>1 are pre-defined and constant during training. It involves a special form of depthwise convolution operation called a Pre-defined Filter Module (PFM). In the channel-wise convolution part, the 1xnxn kernels are drawn from a fixed pool of only a few (16) different pre-defined kernels. In the 1x1 convolution part linear combinations of the pre-defined filter outputs are learned. Despite this harsh restriction, complex and discriminative features are learned. These findings provide a novel perspective on the way how information is processed within deep CNNs. We discuss various properties of PFCNNs and prove their effectiveness using the popular datasets Caltech101, CIFAR10, CUB-200-2011, FGVC-Aircraft, Flowers102, and Stanford Cars. Our implementation of PFCNNs is provided on Github https://github.com/Criscraft/PredefinedFilterNetworks